Apparatus and method to assess the risk of R-on-T event

Abstract

A medical device and a method is suggested for assessing the risk of R on T events. The device comprises a memory, input means for acquiring or receiving an electrogram signal and processing means. The processing means are adapted to detect R-wave and T-waves represented by said electrogram, establish a QT-RR regression model based detected R-waves and T-waves, estimate a vulnerable period, and store estimated vulnerable period data in said memory. Likewise, the method comprises the steps of to detecting R-wave and T-waves represented by an electrogram, establishing a QT-RR regression model based detected R-waves and T-waves, estimating a vulnerable period, and storing estimated vulnerable period data.

1. A medical device comprising:
 a memory; 
 an input configured to receive a cardiac electrogram signal; and, 
 a processor coupled with said memory and said input wherein said processor is configured to:
 detect R-wave and T-waves represented by said cardiac electrogram signal to form one or more binned RR interval range; 
 establish a QT-RR regression model based on said R-waves and T-waves; 
 estimate a vulnerable period to provide estimated vulnerable period data based on a distribution of corresponding RT peak to peak intervals or RT PP  intervals for each said binned RR interval range, to define boundaries of the vulnerable period for said binned RR interval range; 
 store said estimated vulnerable period data in said memory. 
 
 
     
     
       2. The medical device according to  claim 1 , wherein said processor is further configured to establish said QT-RR regression model based on said R-waves and T-waves via tests of different regression models including linear model, hyperbolic model, or parabolic model, with different regression parameters to thus determine an optimal QT-RR regression model, which is defined to have a lowest residuum between modeled data and measured data. 
     
     
       3. The medical device according to  claim 1 , wherein said processor is further configured to conduct said regression analysis offline in an external device, and to program a resulting optimal QT-RR regression model and its parameters into an implantable device. 
     
     
       4. The medical device according to  claim 3 , wherein said processor is further configured to pre-calculate a QT-RR lookup table based on said QT-RR regression model and to download the QT-RR lookup table into said memory. 
     
     
       5. The medical device according to  claim 1 , wherein said processor is further configured to estimate said vulnerable period around a peak of a T-wave through determination of the peak of the T wave and set the vulnerable period around the peak of the T wave that is calculated for each heart cycle based on said QT-RR regression model. 
     
     
       6. The medical device according to  claim 1 , wherein said processor is further configured to define upper and lower boundaries of the vulnerable period for each binned RR interval range as a max/min RT PP  interval. 
     
     
       7. The medical device according to  claim 1 , wherein said processor is further configured to define upper and lower boundaries of the vulnerable period for each binned RR interval so that a programmable percentile of RT PP  intervals are distributed between said upper and lower boundaries of the vulnerable period. 
     
     
       8. A medical device comprising:
 a memory; 
 an input configured to receive a cardiac electrogram signal; 
 a processor coupled with said memory and said input wherein said processor is configured to:
 detect R-wave and T-waves represented by said cardiac electrogram signal to form one or more binned RR interval range; 
 establish a QT-RR regression model based on said R-waves and T-waves; 
 estimate a vulnerable period to provide estimated vulnerable period data; 
 store said estimated vulnerable period data in said memory; 
 calculate a risk score of an R-on-T event RS wherein
   RS=Ae −Δ     2     /σ     2      
 
 with Δ being an absolute time difference between a next RR interval and a nearest boundary of a current beat's vulnerable period if the next RR interval ends up outside vulnerable period boundaries, and being 0 if the next RR interval ends up within the vulnerable period boundaries, A being a constant, that defines a probability of said R-on-T event in a case of Δ=0, and σ that controls a sensitivity of the risk score with respect to a change of Δ, through adjustment of a width of a function curve. 
 
 
     
     
       9. A method for determining the vulnerable period and assessing the risk of R-on-T event, comprising the steps:
 receiving a cardiac electrogram signal on an input; 
 detecting R-waves and T-waves represented by said cardiac electrogram signal and forming one or more binned RR interval range; 
 establishing a QT-RR regression model based on said R-waves and T-waves; 
 estimating a vulnerable period to provide estimated vulnerable period data based on a distribution of the corresponding RT peak to peak intervals or RT PP  intervals for each said binned RR interval range; 
 defining boundaries of the vulnerable period for said binned RR interval range based on said estimating; and, 
 storing said estimated vulnerable period data in a memory of a medical device. 
 
     
     
       10. The method of  claim 9 , wherein the step of establishing a QT-RR regression model comprises:
 testing different regression models including linear model, hyperbolic model, or parabolic model, with different regression parameters; and, 
 determining an optimal QT-RR regression model, which is defined to have a lowest residuum between modelled data and measured data. 
 
     
     
       11. The method of  claim 9 , wherein the step of establishing a QT-RR regression model comprises:
 conducting said regression analysis offline in an external device; and, 
 programming a resulting optimal QT-RR regression model and its parameters into said medical device. 
 
     
     
       12. The method of  claim 11 , wherein the step of establishing a QT-RR regression model comprises:
 pre-calculating a QT-RR lookup table based on said QT-RR regression model; and, 
 downloading said pre-calculated QT-RR lookup table into said memory. 
 
     
     
       13. The method of  claim 9 , wherein the step of estimating the vulnerable period comprises:
 determining a peak of a T wave and then setting the vulnerable period around the peak of the T wave that is calculated for each heart cycle based on said regression model. 
 
     
     
       14. The method of  claim 9 , wherein the step of estimating the vulnerable period further comprises:
 defining upper and lower boundaries of the vulnerable period for each binned RR interval range as a max/min RT PP  interval. 
 
     
     
       15. The method of  claim 9 , wherein the step of estimating the vulnerable period further comprises:
 defining upper and lower boundaries of the vulnerable period for each binned RR interval so that a programmable percentile of the RT PP  intervals are distributed between upper and lower boundaries of the vulnerable period. 
 
     
     
       16. A method for determining the vulnerable period and assessing the risk of R-on-T event, comprising the steps:
 receiving a cardiac electrogram signal on an input; 
 detecting R-waves and T-waves represented by said cardiac electrogram signal and forming one or more binned RR interval range; 
 establishing a QT-RR regression model based on said R-waves and T-waves; 
 estimating a vulnerable period to provide estimated vulnerable period data; 
 storing said estimated vulnerable period data in a memory of a medical device; 
 calculating a risk score of R-on-T event RS wherein;
   RS=Ae −Δ     2     /σ     2      
 
 with Δ being an absolute time difference between a next RR interval and a nearest boundary of a current beat's vulnerable period if the next RR interval ends up outside the vulnerable period boundaries, and being 0 if the next RR interval ends up within the vulnerable period boundaries, A being a constant, that defines a probability of said R-on-T event in a case of Δ=0, and σ controlling a sensitivity of the risk score with respect to a change of Δ, by adjusting a width of a function curve.